出版者・発行元：Proceedings - 2024 IEEE International Conference on Cluster Computing Workshops, CLUSTER Workshops 2024  

For future high-resolution atmospheric simulations, a dynamical core using the discontinuous Galerkin Method (DGM), called SCALE-DG [1], is being developed as an option for high-order fluid schemes in the SCALE library [2]. Compared to the traditional Finite Element Method (FEM), the DGM allows for discontinuities across element boundaries. When evaluating a first-order derivative operator, we use the values at nodes of own element and at common boundaries of neighbor elements. This feature allows most computations to be performed independently for each element. Thus, we can take full advantage of data locality. Additionally, the DGM can achieve high-order accuracy by choosing high-order polynomial basis functions within each element. Therefore, DGM is suitable for high-resolution atmospheric simulations with high-order numerical accuracy, and we expect the computational performance to be highly desirable on modern computer architectures.

DOI： 10.1109/CLUSTERWORKSHOPS61563.2024.00033

Web of Science

Scopus

出版者・発行元：Proceedings - 2024 IEEE 17th International Symposium on Embedded Multicore/Many-core Systems-on-Chip, MCSoC 2024  

The unregulated utilization of Artificial Intelligence (AI) outputs, potentially leading to various societal issues, has received considerable attention. While humans routinely validate information, manually inspecting the vast volumes of AI-generated results is impractical. Therefore, automation and visualization are imperative. In this context, Explainable AI (XAI) technology is gaining prominence, aiming to streamline AI model development and alleviate the burden of explaining AI outputs to users. Simultaneously, software Auto-Tuning (AT) technology has emerged for reducing the man-hours required for performance tuning in numerical calculations. AT is a potent tool for cost reduction during parameter optimization and high-performance programming for numerical computing. The synergy between AT mechanisms and AI technology is noteworthy, with AI finding extensive applications in AT. However, applying AI to AT mechanisms introduces challenges in AI model explainability. This study focuses on XAI for AI models when integrated into two different processes for practical numerical computations: performance parameter tuning of accuracy-guaranteed numerical calculations and sparse iterative algorithm.

DOI： 10.1109/MCSoC64144.2024.00095

Scopus

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Proceedings - 2023 11th International Symposium on Computing and Networking, CANDAR 2023  

Mixed-precision computation is a promising method for substantially increasing the speed of numerical computations. However, using mixed-precision data is a double-edged sword. Although it can improve the computational performance, the reduction in precision brings more uncertainties and errors. It is necessary to determine which variables can be represented with a lower-precision format without affecting the accuracy of the results. Hence, much effort is spent on selecting appropriate variables while considering the execution time and numerical accuracy. Auto-Tuning (AT) is one of several technologies that can assist in eliminating this intensive work. In this study, we investigated an AT strategy for the 'Blocks' directive in the auto-Tuning language ppOpen-AT to tune multiple regions of a program and evaluated the effectiveness. A benchmark program of the nonhydrostatic icosahedral atmospheric model (NICAM), which is a global cloud resolving model, was considered as a study case. Experimental results indicated that when a single part of the program could perform well in the mixed-precision computation, a combination achieved a better performance. When used on the Flow Type I Subsystem (The Fujitsu PRIMEHPC FX1000), this method achieved almost 1.27× speedup compared with the NICAM benchmark program using all double-precision data.

DOI： 10.1109/candar60563.2023.00031

Scopus

記述言語：日本語   出版者・発行元：日本医用画像工学会  

<p>医用画像処理技術の発達により，生体の内部を視覚的に理解するためのさまざまな技術が開発され，利用されている．しかし，それらにより直接的に得ることができるのは画像や映像であり，診断は医師など人の手によって行われている．これらの労力を軽減するソフトウェアへの期待は大きく，すでに医療の現場で利用されている技術も増えてきているが，医療（医用画像）と計算機技術の両方の知識と技術が必要なため，対象は限られている．そこで本研究では，医用画像処理分野と高性能計算分野の研究者が協力してPET における画像再構成の高速化と大規模化に取り組んでいる．本稿ではその取り組みの内容とこれまでに得られた成果を紹介する．</p>

DOI： 10.11409/mit.41.150

CiNii Research

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：2023 IEEE International Parallel and Distributed Processing Symposium Workshops, IPDPSW 2023  

With the NVIDIA Turing architecture generation, several NVIDIA graphics processing units (GPUs) have introduced ray tracing acceleration hardware (RT cores). Ray tracing processing can be regarded as a simulation of wave and particle propagation, collision, and reflection. Therefore, it is expected to be applied to computational science and high-performance computing. However, few studies have been conducted using RT cores. The purpose of this research is to demonstrate the use of RT cores in the scientific and technical computing fields. We implemented a radio wave propagation loss calculation with the programmable ray tracing application framework OptiX and evaluated its performance. Furthermore, we investigated the challenges of reducing the description of framework-specific settings and the needs of hardware allocation. In the simple two spheres experiment, the RT core implementation showed the highest performance. Moreover, the acceleration was super linear scaling, between (10000, 5000) and (20000, 10000). In the experiment with a sphere and planes, the performance achieved by the RT cores was up to approximately 390 times higher than the parallel execution of the BVH search algorithm. We also proved that a large number of RT cores yielded higher performance. In the open data problem space experiment, we evaluated various GPUs and revealed that a larger number of RT cores is effective. These results show that RT cores are sufficiently effective for radio propagation calculations with an adequate number of ray projections. Through this research, we contributed to the RT core use in computational science by proposing an implementation method for ray tracing applications and revealing the effects of RT cores in radio wave propagation loss calculations.

DOI： 10.1109/IPDPSW59300.2023.00115

Web of Science

Scopus

その他リンク： https://dblp.uni-trier.de/db/conf/ipps/ipdps2023w.html#HashinokiOKNH23

掲載種別：研究論文（学術雑誌）   出版者・発行元：Oxford University Press (OUP)  

Summary

Large-scale earthquake sequence simulations using the boundary element method (BEM) incur extreme computational costs through multiplying a dense matrix with a slip rate vector. Hierarchical matrices (H-matrices) have often been used to accelerate this multiplication. However, the complexity of the structures of the H-matrices and the communication costs between processors limit their scalability, and they therefore cannot be used efficiently in distributed memory computer systems. Lattice H-matrices have recently been proposed as a tool to improve the parallel scalability of H-matrices. In this study, we developed a method for earthquake sequence simulations applicable to 3D nonplanar faults with lattice H-matrices. We present a simulation example and verify the mesh convergence of our method for a 3D nonplanar thrust fault using rectangular and triangular discretizations. We also performed performance and scalability analyses of our code. Our simulations, using over ${10^5}$ degrees of freedom, demonstrated a parallel acceleration beyond ${10^4}$ MPI processors and a &amp;gt; 10-fold acceleration over the best performance when the normal H-matrices are used. Using this code, we can perform unprecedented large-scale earthquake sequence simulations on geometrically complex faults with supercomputers. The software is made an open-source and freely available.

DOI： 10.1093/gji/ggac386

arXiv

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

DOI： 10.1109/CLUSTER51413.2022.00056

その他リンク： https://dblp.uni-trier.de/db/conf/cluster/cluster2022.html#HoshinoIH22

CiNii Research

CiNii Research

CiNii Research

CiNii Research

掲載種別：研究論文（国際会議プロシーディングス）  

DOI： 10.1007/s11548-018-1766-y

記述言語：日本語  

J-GLOBAL

記述言語：英語   掲載種別：研究論文（学術雑誌）  

DOI： 10.1007/978-3-319-93698-7_46

Scopus

掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE Computer Society  

DOI： 10.1109/CLUSTER.2018.00016

記述言語：日本語  

J-GLOBAL

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）  

記述言語：日本語   掲載種別：研究論文（学術雑誌）  

J-GLOBAL

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

Directive-based programming interfaces such as OpenACC and OpenMP are becoming more prevalent in application development targeting accelerators, in particular when porting existing CPU-only code. Unlike vendor-specific alternatives such as CUDA, they are designed to be portable across different accelerators, and therefore once necessary directives are added to an existing CPU-only code, it can be executed on different accelerator architectures depending on the availability of supporting compilers. However, it does not automatically mean that such code runs efficiently on different architectures, and in fact, architecture-specific coding such as choosing optimal data layouts is almost mandatory for optimal performance, imposing a significant burden if implemented manually. Towards realizing performance portability in accelerator programming, we propose a set of extended directives that allow the programmer to optimize data layouts for a given accelerator without modifying original program code. Unlike the manual approach, the code change is confined in the directives with the original code kept as it is. This paper evaluates the effectiveness of our proposed extensions in the OpenACC standard by extending UPACS and CCS-QCD OpenACC applications. A prototype source-to-source translator for the extensions achieves 123% and 120% of the baseline performance, respectively, which are comparable to manually tuned versions.

DOI： 10.1109/HPCC-SmartCity-DSS.2016.34

Web of Science

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：Institute of Electrical and Electronics Engineers Inc.  

OpenACC is gaining momentum as an implicit and portable interface in porting legacy CPU-based applications to heterogeneous, highly parallel computational environment involving many-core accelerators such as GPUs and Intel Xeon Phi. OpenACC provides a set of loop directives similar to OpenMP for the parallelization and also to manage data movement, attaining functional portability across different heterogeneous devices
however, the performance portability of OpenACC is said to be insufficient due to the characteristics of different target devices, especially those regarding memory layouts, as automated attempts by the compilers to adapt is currently difficult. We are currently working to propose a set of directives to allow compilers to have better semantic information for adaptation
here, we particularly focus on data layout such as Structure of Arrays, advantageous data structure for GPUs, as opposed to Array of Structures, which exhibits good performance on CPUs. We propose a directive extension to OpenACC that allows the users to flexibility specify optimal layouts, even if the data structures are nested. Performance results show that we gain as much as 96 % in performance for CPUs and 165% for GPUs compared to programs without such directives, essentially attaining both functional and performance portability in OpenACC.

DOI： 10.1109/WACCPD.2014.12

Scopus

記述言語：英語   掲載種別：研究論文（国際会議プロシーディングス）   出版者・発行元：IEEE  

OpenACC is a new accelerator programming interface that provides a set of OpenMP-like loop directives for the programming of accelerators in an implicit and portable way. It allows the programmer to express the offloading of data and computations to accelerators, such that the porting process for legacy CPU-based applications can be significantly simplified. This paper focuses on the performance aspects of OpenACC using two microbenchmarks and one real-world computational fluid dynamics application. Both evaluations show that in general OpenACC performance is approximately 50% lower than CUDA. However, for some applications it can reach up to 98% with careful manual optimizations. The results also indicate several limitations of the OpenACC specification that hamper full use of the GPU hardware resources, resulting in a significant performance gap when compared to a highly tuned CUDA code. The lack of a programming interface for the shared memory in particular results in as much as three times lower performance.

DOI： 10.1109/CCGrid.2013.12

Web of Science

J-GLOBAL

記述言語：日本語   出版者・発行元：小宮山印刷工業  

CiNii Books

記述言語：日本語   出版者・発行元：日本計算工学会  

J-GLOBAL

J-GLOBAL

J-GLOBAL

J-GLOBAL

担当区分：筆頭著者  

J-GLOBAL

J-GLOBAL

J-GLOBAL

記述言語：日本語   出版者・発行元：日本計算工学会  

J-GLOBAL

記述言語：英語  

記述言語：日本語  

J-GLOBAL

記述言語：日本語  

J-GLOBAL

記述言語：日本語  

J-GLOBAL

記述言語：日本語  

J-GLOBAL

記述言語：日本語  

J-GLOBAL

記述言語：日本語  

J-GLOBAL

記述言語：日本語  

J-GLOBAL

記述言語：日本語  

J-GLOBAL

記述言語：日本語   出版者・発行元：一般社団法人情報処理学会  

近年増加傾向にある GPU 等のアクセラレータを搭載した計算環境への既存プログラムの移植方法として，CUDA・OpenCL に代表されるローレベルなプログラミングモデルを用いる方法に対し，ディレクティブベースの OpenACC のようなハイレベルなプログラミングモデルを用いる方法が注目されている．このようなディレクティブベースのプログラミングモデルの利点として，元のプログラムを維持したまま移植を行えるために，デバイス間の機能的な可搬性が高いことがあげられる．しかし現状の OpenACC などの High-level なプログラミングモデルは，スカラプロセッサとメニーコアアクセラレータの得意とするデータレイアウトの相違に対応することが出来ず，異なる性質を持ったデバイス間の性能可搬性に問題がある．そこで本研究では，データレイアウトを抽象化し，異なるデバイス間での性能可搬性を向上させるための OpenACC の拡張ディレクティブを試作し，姫野ベンチマークのデータレイアウトをトランスレーターにより変更し，マルチコア CPU，Intex Xeon Phi，K20X GPU のそれぞれで評価を行った．その結果，オリジナルと同一のデータレイアウトと比較して，Intel Xeon Phi では 27%，K20X GPU では 24%の性能向上が得られることを確認した．

CiNii Books

記述言語：日本語   出版者・発行元：一般社団法人情報処理学会  

近年増加傾向にある GPU 等のアクセラレータを搭載した計算環境への既存プログラムの移植方法として，CUDA・OpenCL に代表される Low-level なプログラミングモデルを用いる方法に対し，ディレクティブベースの OpenACC のような High-level なプログラミングモデルを用いる方法が考えられる．このようなディレクティブベースのプログラミングモデルの利点として，元のプログラムを壊さずに移植を行えるために，デバイス間の可搬性が高いことがあげられる．しかし現状の OpenACC などのプログラミングモデルは，スカラプロセッサとメニーコアアクセラレータの得意とするデータレイアウトの相違等に対応することが出来ず，異なる性質を持ったデバイス間の性能可搬性に問題がある．そこで本研究では，データレイアウトを抽象化し，異なるデバイス間での性能可搬性を向上させるための OpenACC の拡張ディレクティブを試作し，評価を行った．

CiNii Books

記述言語：日本語  

記述言語：日本語  

地震や気象予測，航空機や高層ビル設計といったシミュレーションに利用される数値流体力学アプリケーションは，近年一般的になりつつある GPU を用いたスーパーコンピュータにおいて，目覚ましい成果を上げている．しかし，GPU を用いたプログラミングは，高い性能を得ること難しいと言われており，レガシープログラムの GPU 環境への移植が問題となっている．本稿では，実際に利用されている大規模流体アプリケーションである UPACS を手動により CUDA 化し，性能と移植コストの面から評価を行った．また，プログラムの移植性を解決すると期待されている，OpenACC の予備評価を行った．これら評価の結果を示し，今後解決すべき課題について述べる．Computational fluid dynamics (CFD) applications used for an earthquake and meteorological simulation are one of the most important application executed with high-speed supercomputers. Especially, GPU-based supercomputers have been showing remarkable performance of CFD applications. However, GPU-programing is still difficult to obtain high performance, which prevents legacy applications from being ported to GPU environment. We apply classical optimizations to a real-world CFD application UPACS and evaluate it&#039;s performance and porting costs, and we also evaluate OpenACC expected to provide portability across CPUs and GPUs. We demonstrate these results of evaluation and mention performance problems should be resolved in the future.

CiNii Books

記述言語：日本語  

記述言語：英語   出版者・発行元：一般社団法人映像情報メディア学会  

DOI： 10.3169/itej.66.817

Scopus

CiNii Books